Colab now has a Paid Subscription Option

https://colab.research.google.com/signup

Below I have quotes from the FAQ. The usage limit concerns me because what is “too much?” A language model is high usage (in terms of GPU) and needs to train At least a few hours. Would I be put in this group? I do not know. From what it sounds like the free version (can) get access to T4’s and P100’s, but most of the time they’ll get K80’s. Which isn’t the end of the world.

Some quotes:

What kinds of GPUs are available in Colab Pro?

With Colab Pro you get priority access to our fastest GPUs. For example, you may get access to T4 and P100 GPUs at times when non-subscribers get K80s. You also get priority access to TPUs. There are still usage limits in Colab Pro, though, and the types of GPUs and TPUs available in Colab Pro may vary over time.

In the free version of Colab there is very limited access to faster GPUs, and usage limits are much lower than they are in Colab Pro.

How long can notebooks run in Colab Pro?

With Colab Pro your notebooks can stay connected for up to 24 hours, and idle timeouts are relatively lenient. Durations are not guaranteed, though, and idle timeouts may sometimes vary.

In the free version of Colab notebooks can run for at most 12 hours, and idle timeouts are much stricter than in Colab Pro.

How much memory is available in Colab Pro?

With Colab Pro you get priority access to high-memory VMs. These VMs generally have double the memory of standard Colab VMs, and twice as many CPUs. You will be able to access a notebook setting to enable high-memory VMs once you are subscribed. Additionally, you may sometimes be automatically assigned a high-memory VM when Colab detects that you are likely to need it. Resources are not guaranteed, though, and there are usage limits for high memory VMs.

In the free version of Colab the high-memory preference is not available, and users are rarely automatically assigned high memory VMs.

Why aren’t resources guaranteed in Colab Pro?

In order to offer faster GPUs, longer runtimes and more memory in Colab for a relatively low price, Colab needs to maintain the flexibility to adjust usage limits and hardware availability on the fly. Although we make no guarantees, we anticipate that most subscribers who use Colab Pro as it is intended to be used – for interactive computing – will experience few if any usage limits.

How can I get the most out of Colab Pro?

Resources in Colab Pro are prioritized for subscribers who have recently used less resources, in order to prevent the monopolization of limited resources by a small number of users. To get the most out of Colab Pro, consider closing your Colab tabs when you are done with your work, and avoid opting for GPUs or extra memory when it is not needed for your work. This will make it less likely that you will run into usage limits within Colab Pro. For more information, see Taking Advantage of Colab Pro

For a fair comparison, Paperspace Gradient has a $8/month subscription which (among other GPU’s) includes a P4000 and P5000

Along with this is a “configurable auto shutdown”.

https://docs.paperspace.com/gradient/instances/instance-tiers

So while Colab has higher GPU’s available (but not guaranteed to get the best ones) you still have a 24 hour limit (and persistent storage is still google drive)

Only concerning bit is the free users may not get a GPU some times (of Colab)

1 Like

For now, Colab Pro is only available in the US.

Can anyone confirm this? If someone out of US has it, does it work well?

IIRC I saw that people can set their address to somewhere in the US and it works. You didn’t hear that from me though :eyes:

2 Likes